son·o·lu·mi·nes·cence (sn-lm-nsns)--"The production of light as a result of the passing of sound waves through a liquid medium. The sound waves cause the formation of bubbles that emit bright flashes of light when they collapse." http://www.youtube.com/watch?v=GJolILUbdNw
The Wikipedia page has a list of several possible explanations: http://en.wikipedia.org/wiki/Mechanism_of_sonoluminescence
I'm liking the hot spot theory. The temperatures inside that bubble can be incredibly high, resulting perhaps in the brief production of a plasma gas. "Because the bubble collapse occurs within microseconds,[5] the hot spot theory states that the thermal energy results from an adiabatic bubble collapse. In 1950 it was assumed that the bubble internal temperatures were as high as 10,000 K at the collapse of a spherical symmetric bubble.[1] In the 1990s, sonoluminescence spectra were used by Suslick to measure effective emission temperatures in bubble clouds (multibubble sonoluminescence) of 5000 K,[12][13] and more recently temperatures as high as 20,000 K in single bubble cavitation." Please Register or Log in to view the hidden image!
I've seen it in action. Centrifugal pumps are frequently used in refining process. Losing 'minimum' suction head pressure can result in cavitation. Serious cavitation where the exploding bubbles rock the pump foundation or crack the suction line or pump case depending where the cavitation is centered. These are extreme results but 'short periods' of cavitation can heat up the suction and pump case, blow the pump seal, and threaten hydrocarbon fire. The case and suction line will be smoking hot. Not surprised to learn light emission can happen in this extreme environment. Another interesting link http://www.nature.com/news/2005/050228/full/news050228-7.html One of the many things learned through the SOHO experiment is how much sound waves, generated inside the sun, effect the temperature outside the surface region of the sun. The mechanical energy of the sound waves is conserved as heat energy when the mechanical waves no longer can propagate. That's a great experiment to follow.
Just thought while we're on this topic, I should mention that there are many materials which convert light directly into sound. Alexander Graham Bell experimented with such materials over 130 years ago when he was developing his most prized invention, the "photophone", an early precursor to fibre optics and radio. One such substance was known as "lampblack", some kind of powdery substance left over from burning petroleum jelly. In the earliest versions of the photophone, all Bell had to do was stick some of this stuff up against his eardrum, and even the noise produced from ordinary sunlight was enough to be painful. Ultimately Bell developed an electrical device to convert the photophone's light into sound, but I find it neat that he was also able to achieve this effect and produce a coherent sound with nothing but some powder and a source of light.